{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:5HTNLYAVM6X6EFF2MX634K2GZS","short_pith_number":"pith:5HTNLYAV","schema_version":"1.0","canonical_sha256":"e9e6d5e01567afe214ba65fdbe2b46cc8773f3ac24fa4fe22897cb765f322506","source":{"kind":"arxiv","id":"2505.24201","version":1},"attestation_state":"computed","paper":{"title":"SentinelAgent: Graph-based Anomaly Detection in Multi-Agent Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Di Wu, Kun Sun, Xu He, Yan Zhai","submitted_at":"2025-05-30T04:25:19Z","abstract_excerpt":"The rise of large language model (LLM)-based multi-agent systems (MAS) introduces new security and reliability challenges. While these systems show great promise in decomposing and coordinating complex tasks, they also face multi-faceted risks across prompt manipulation, unsafe tool usage, and emergent agent miscoordination. Existing guardrail mechanisms offer only partial protection, primarily at the input-output level, and fall short in addressing systemic or multi-point failures in MAS. In this work, we present a system-level anomaly detection framework tailored for MAS, integrating structu"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.24201","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-05-30T04:25:19Z","cross_cats_sorted":[],"title_canon_sha256":"eca03de1c9f81b972f80c21d3f4ddf62e2a889eebff68a91f414d6de38c25a38","abstract_canon_sha256":"f8b92379d04323e9c8a9fc0eef2d54e1019e5a2eb1182760eec61e15d7fcb028"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:43.933190Z","signature_b64":"kyrZSbZ9R5VBotLeUKPSTA+pbm3wca0bBBeW+2wNfLS+fkiysXD8sK77lHYi0DJdfgB/qpLLqr87M9aCGRNCCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e9e6d5e01567afe214ba65fdbe2b46cc8773f3ac24fa4fe22897cb765f322506","last_reissued_at":"2026-07-05T11:12:43.932631Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:43.932631Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SentinelAgent: Graph-based Anomaly Detection in Multi-Agent Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Di Wu, Kun Sun, Xu He, Yan Zhai","submitted_at":"2025-05-30T04:25:19Z","abstract_excerpt":"The rise of large language model (LLM)-based multi-agent systems (MAS) introduces new security and reliability challenges. While these systems show great promise in decomposing and coordinating complex tasks, they also face multi-faceted risks across prompt manipulation, unsafe tool usage, and emergent agent miscoordination. Existing guardrail mechanisms offer only partial protection, primarily at the input-output level, and fall short in addressing systemic or multi-point failures in MAS. In this work, we present a system-level anomaly detection framework tailored for MAS, integrating structu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24201","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2505.24201/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.24201","created_at":"2026-07-05T11:12:43.932699+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24201v1","created_at":"2026-07-05T11:12:43.932699+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24201","created_at":"2026-07-05T11:12:43.932699+00:00"},{"alias_kind":"pith_short_12","alias_value":"5HTNLYAVM6X6","created_at":"2026-07-05T11:12:43.932699+00:00"},{"alias_kind":"pith_short_16","alias_value":"5HTNLYAVM6X6EFF2","created_at":"2026-07-05T11:12:43.932699+00:00"},{"alias_kind":"pith_short_8","alias_value":"5HTNLYAV","created_at":"2026-07-05T11:12:43.932699+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22741","citing_title":"GRADE: Graph Representation of LLM Agent Dependency and Execution","ref_index":119,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01581","citing_title":"Agent System Operations: Categorization, Challenges, and Future Directions","ref_index":107,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30602","citing_title":"MESA: Prioritizing Vulnerable Communication Channels for Securing Multi-Agent Systems","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11053","citing_title":"Content-Aware Attack Detection in LLM Agent Tool-Call Traffic: An Empirical Study of Features, Architectures, and Evaluation Protocols","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23414","citing_title":"When Planning Fails Despite Correct Execution: On Epistemic Calibration for LLM-Based Multi-Agent Systems","ref_index":81,"is_internal_anchor":false},{"citing_arxiv_id":"2510.21236","citing_title":"AgentBound: Securing Execution Boundaries of AI Agents","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2603.04474","citing_title":"From Spark to Fire: Modeling and Mitigating Error Cascades in LLM-Based Multi-Agent Collaboration","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12856","citing_title":"Moltbook Moderation: Uncovering Hidden Intent Through Multi-Turn Dialogue","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11053","citing_title":"Content-Aware Attack Detection in LLM Agent Tool-Call Traffic: An Empirical Study of Features, Architectures, and Evaluation Protocols","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12856","citing_title":"Moltbook Moderation: Uncovering Hidden Intent Through Multi-Turn Dialogue","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2604.01151","citing_title":"Detecting Multi-Agent Collusion Through Multi-Agent Interpretability","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11053","citing_title":"Content-Aware Attack Detection in LLM Agent Tool-Call Traffic: An Empirical Study of Features, Architectures, and Evaluation Protocols","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14317","citing_title":"Challenges and Future Directions in Agentic Reverse Engineering Systems","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS","json":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS.json","graph_json":"https://pith.science/api/pith-number/5HTNLYAVM6X6EFF2MX634K2GZS/graph.json","events_json":"https://pith.science/api/pith-number/5HTNLYAVM6X6EFF2MX634K2GZS/events.json","paper":"https://pith.science/paper/5HTNLYAV"},"agent_actions":{"view_html":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS","download_json":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS.json","view_paper":"https://pith.science/paper/5HTNLYAV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24201&json=true","fetch_graph":"https://pith.science/api/pith-number/5HTNLYAVM6X6EFF2MX634K2GZS/graph.json","fetch_events":"https://pith.science/api/pith-number/5HTNLYAVM6X6EFF2MX634K2GZS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS/action/storage_attestation","attest_author":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS/action/author_attestation","sign_citation":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS/action/citation_signature","submit_replication":"https://pith.science/pith/5HTNLYAVM6X6EFF2MX634K2GZS/action/replication_record"}},"created_at":"2026-07-05T11:12:43.932699+00:00","updated_at":"2026-07-05T11:12:43.932699+00:00"}